Abstract: The invention notably relates to a computer-implemented method for designing a mechanical part, the method comprising: providing a subset of a finite element mesh (FEM), the subset of the FEM representing the mechanical part; and determining a representation of a skeleton of the mechanical part based on the subset of the FEM, the skeleton having branches and branch junctions, each branch junction joining respective branches. This improves the designing of a mechanical part.

Abstract: The invention notably relates to a computer-implemented method for retrieving model outcomes in the course of an event. The method comprises providing variables of a first set of observations of similar events, providing a multiplicity of models, indexing the multiplicity of models with the variables of the first set of observations, querying a model according to one or more variables, and returning, as a result of the query, a model.

Abstract: The present invention relates to a method and corresponding system for generating a computer-aided design (CAD) model from a finite element mesh. The method of the invention begins with selecting one or more mesh-element-faces on a finite element mesh that represents a geometric object to be formed of one or more geometric faces. Next, from the one or more selected mesh-element-faces, respective geometric faces are generated. Finally, any generated geometric faces are stitched together to make a geometric shell of the geometric object that the finite element mesh represents.

Abstract: The invention notably relates to a computer-implemented method for querying a database. The database comprises 3D modeled objects that each represent a mechanical part. The database further comprises, for each 3D modeled object, a respective morphology signature. The method comprising: providing a query that includes a morphology criterion; returning, as results of the query, respective 3D modeled objects of the database, a respective 3D modeled object being returned based on an extent to which the respective 3D modeled object has a morphology signature that respects the morphology criterion. Such a method improves the field of searching mechanical parts in a database.

Abstract: A computer-implemented method of signal processing comprises providing images. The method comprises for each respective one of at least a subset of the images: applying a weakly-supervised learnt function, the weakly-supervised learnt function outputting respective couples each including a respective localization and one or more respective confidence scores, each confidence score representing a probability of instantiation of a respective object category at the respective localization. The method further comprises determining, based on the output of the weakly-supervised learnt function, one or more respective annotations, each annotation including a respective localization and a respective label representing instantiation a respective object category at the respective localization. The method further comprises forming a dataset including pieces of data, each piece of data including a respective image of the subset and at least a part of the one or more annotations determined for the respective image.

Abstract: A computer-implemented method of signal processing comprises providing images. The method comprises for each respective one of at least a subset of the images: applying a weakly-supervised learnt function, the weakly-supervised learnt function outputting respective couples each including a respective localization and one or more respective confidence scores, each confidence score representing a probability of instantiation of a respective object category at the respective localization. The method further comprises determining, based on the output of the weakly-supervised learnt function, one or more respective annotations, each annotation including a respective localization and a respective label representing instantiation a respective object category at the respective localization. The method further comprises forming a dataset including pieces of data, each piece of data including a respective image of the subset and at least a part of the one or more annotations determined for the respective image.

Abstract: An example embodiment designs a real-world object by defining a first model of the object being produced using an additive manufacturing (AM) process, where behavior of the object being produced is given by a first equation which includes a first plurality of corresponding sensitivity equations for a first plurality of design variables. Similarly, such an embodiment defines a second model of the object after being produced, wherein behavior of the object after being produced is given by a second equation which includes a second plurality of corresponding sensitivity equations for a second plurality of design variables. In turn, the second model is iteratively optimized with respect to a given one of the second plurality of design variables using both the first plurality of corresponding sensitivity equations and the second plurality of corresponding sensitivity equations.

Abstract: Embodiments automatically determine optimized designs for manufacturing real-world objects. An embodiment begins with defining a finite element model comprised of a plurality of elements that represents a real-world object. Next, equilibriums and design responses of the object in response boundary conditions are determined, which includes calculating a local volume constraint for a given element of the finite element model. Then, design response sensitivities of the object in response to the boundary conditions are determined, which includes differentiating the calculated local volume constraint to determine sensitivity of a sizing design variable.

Abstract: A method for simulation, in particular for computerized calculation, of at least one physical property of a system of one or more chemical species that includes at least one chemical species dissolved in at least one inhomogeneously distributed chemical species, using a quasi-chemical calculation based on the statistical thermodynamics of pairwise interactions of molecule surface segments. According to the invention, pressure, that arises from the statistical thermodynamic over- or underpopulation of spatial regions, interacts as an additional continuum response function with the atomic volumes and thus influences the thermodynamic weight of a molecular state in the system, during the iterative calculation of the statistical thermodynamic distribution of molecules in 1-dimensionally, 2-dimensionally, or 3-dimensionally structured simulation volumes of liquid systems.

Abstract: Unlike existing methods that rely on manual procedures for repairing finite element meshes in computer-based simulations, embodiments automatically repair finite element meshes for use in simulations of real-world objects. One such embodiment begins by identifying a non-compliant mesh element in a finite element mesh and extracting a mesh patch from the finite element mesh that includes the identified non-compliant mesh element. To continue, an invariant patch description for the extracted mesh patch is generated and a repair solution corresponding to the generated invariant patch description is obtained from a database storing pre-determined repair solutions. In turn, the mesh patch in the finite element mesh is repaired using the obtained repair solution.

Abstract: A computer-implemented method for designing, with a CAD system, a 3D modeled object representing a mechanical part. The method includes displaying a B-rep having faces and edges and representing at least a part of the mechanical part, selecting, by graphical user interaction, at least one face of the B-rep, and, automatically by the CAD system, recognizing a set of faces including the at least one face and representing a bump. The recognizing includes determining a first set of faces representing a depression and a second set of faces representing a protrusion, and selecting, among the first set of faces and the second set of faces, a set of faces having a shortest boundary curve perimeter and being different from both the selected at least one face and the B-rep. The first set of faces and the second set of faces both comprise the at least one face.

Abstract: The disclosure notably relates to a computer-implemented method for teaching a generative autoencoder. The generative autoencoder is configured to generate functional structures. A functional structure is a data structure representing a mechanical assembly of rigid parts and which includes a tree. Each leaf node represents a shape and positioning of a respective rigid part and a force exerted on the respective rigid part. Each non-leaf node with several children represents a mechanical link between sub-assemblies. Each sub-assembly is represented by a respective one of the several children. Each non-leaf node with a single child represents a duplication of the sub-assembly represented by the single child. The method includes obtaining a dataset including functional structures. The method further comprises teaching the generative autoencoder on the dataset. This constitutes an improved method for teaching a generative autoencoder configured for generating functional structures.

Abstract: A computer-implemented method designs and manufactures a supporting structure for the packaging of a solid object. The supporting structure comprises a plurality of linear support elements that sustain the solid object inside a packaging container. The method includes: a) providing, as an input, a three-dimensional model of the solid object; b) computing a first cumulative linear mass density distribution of the solid object according to a first axis (x); and c) using said first cumulative linear mass density distribution to determine the positions, along said first axis, of linear support elements oriented transversally to said first axis. The resultant positions enable even distribution of the weight of the solid object among the linear support elements. A computer program product, computer-readable data-storage medium, and CAD system carry out such a method.

Abstract: A computer-implemented method of generating and displaying an exploded view of a three-dimensional model of an assembly (A) of parts (01, 02, 03), the three-dimensional model having a hierarchical structure comprising a plurality of levels, the method comprising the steps of: a) displaying an unexploded view of the assembly; b) displaying a graphical tool (GT) representing the levels of the structure of the assembly and, for each level, an explosion ratio; c) receiving an input for selecting a level and an explosion ratio, and using the graphical tool for displaying them; d) displaying an exploded view of the assembly, wherein only groups of parts associated to higher levels than the selected one are taken apart. A computer program product, non-volatile computer-readable data-storage medium and computer system for carrying out such a method.

Abstract: A method comprising: simulating, in a lattice velocity set, movement of particles in a volume of fluid, with the movement causing collision among the particles; based on the simulated movement, determining relative particle velocity of a particle at a particular location within the volume, with the relative particle velocity being a difference between (i) an absolute velocity of the particle at the particular location within the volume and measured under zero flow of the volume, and (ii) a mean velocity of one or more of the particles at the particular location within the volume; and determining, based on the relative particle velocity, a non-equilibrium post-collide distribution function of a specified order that is representative of the collision.

Abstract: The present disclosure is directed to a method and corresponding system that improves accuracy of a computer simulation of an original posture of a digital human model (DHM) relative to a target object. The method and system may obtain information associated with the original DHM posture. The obtained DHM posture information may include a position of a head of the DHM. The method and system may obtain information associated with the target object. The obtained target object information may include a size of the target object and an orientation of the target object. The method and system method may obtain a distance from the head of the DHM to the target object. In some embodiments, the system and method may generate a measure of vision (i.e., vision measure) of the DHM of the target object that the DHM is visually targeting.

Abstract: A computer-implemented method is provided for simulating a modal frequency response of a real-world object. The computer-implemented method includes dividing a plurality of excitation frequencies into a plurality of excitation frequency subsets, calculating modal frequency responses for at least a portion of the excitation frequencies in a given excitation frequency subset, and generating a simulation of the real-world object based at least in part on the modal frequency responses.

Abstract: The disclosure notably relates to a computer-implemented method for designing an outer surface of a composite part manufactured by molding a stack of material layers. The method includes defining constant offset surfaces, a constant offset surface being a respective part of the outer surface which is to have a constant offset value relative to the reference surface, the constant offset value of a respective constant offset surface corresponding to the sum of the thicknesses of the material layers below the respective constant offset surface, and determining a final surface that corresponds to a tangent continuous connection of the constant offset surfaces. This provides an improved solution for designing an outer surface of a composite part.

Abstract: A computer-implemented method for learning an autoencoder notably is provided. The method comprises providing a dataset of images. Each image includes a respective object representation. The method also comprises learning the autoencoder based on the dataset. The learning includes minimization of a reconstruction loss. The reconstruction loss includes a term that penalizes a distance for each respective image. The penalized distance is between the result of applying the autoencoder to the respective image and the set of results of applying at least part of a group of transformations to the object representation of the respective image. Such a method provides an improved solution to learn an autoencoder.

Abstract: A computer-implemented method for manipulating at least one three-dimensionally modeled object (O1) in a three-dimensional scene displayed on a computer display. The method displays, in the three-dimensional scene, a cursor (C) controlled by a pointing device (PD). The cursor has a shape representing three axes (X, Y, Z) of a coordinate system. Next the method selects an object of the three-dimensional scene; and detects a drag operation performed by a user using the pointing device. The drag operation translates the cursor and selects an axis of the coordinate system as a function of the translation of the cursor. Also the method performs a manipulation of the selected object. The manipulation depends on the selected axis and on a length (L) of the translation of the cursor (PP) along the selected axis.